In this second post about doubts about ELF pronunciation I want to respond to two questions about weak forms:
1. … regarding the use of weak forms, the LFC states that they are not to be taught (unless the student’s needs are for EFL) so speakers are encouraged to use the strong form of the word. In the cases of “a” and “the” (which are very common words in English), the strong forms (/eɪ/ and /ðɪ/) are far from the weak forms (/ə/ and /ðə/, respectively). What should be an option to learn in those cases?
2. My last question regarding the use of weak forms in ELF is in relation to the vowels that occur in the strong forms. For instance, in the word “that”, the strong form in RP is /ðæt/. Nevertheless, the vowel in that word is difficult for Spanish speakers to pronounce, so they would substitute it for /a/. Would that cause an intelligibility problem?
With respect to the first question, an ELF approach certainly does not encourage the teaching of weak form pronunciations in terms of productive competence because of the negative impact they can have on intelligibility. This is something that runs completely counter to an EFL approach to pronunciation, and to the (at times) almost obsessive focus these last 30 years on connected speech and the different modifications that native speakers make beyond a certain minimum speed of speech.
However, anyone who is familiar with Richard Cauldwell’s excellent Phonology for Listening (Speech in Action, 2013. http://www.speechinaction.org/ebooks/) will know that it is precisely the multiple modifications to the citation forms of words that native speakers make, that make them so difficult to understand for many nonnative speaker listeners. Because of this, in an ELF setting where interlocutors are working collaboratively towards maximum intelligibility, weak forms and certain other connected speech modifications are counter-productive.
What is not correct in Q1, however, is to say that the LFC encourages ELF speakers to use the strong form. Avoiding the weak form does not automatically mean striving for a NS strong-form vowel as the alternative, and the key is to get learners to focus on the correct placement of nuclear stress (a key item in the LFC), which means ‘de-focusing’ the unstressed syllables in the tone unit.
In practice with my own students I find that what comes out in terms of vowel quality is largely governed by the written vowel on the page, but the exact quality of vowels in not an issue in ELF pronunciation (provided a speaker’s vowel qualities are consistent). None of my Spanish-L1 students, for example, would produce /eɪ/ or /ðɪ/ for ‘a’ or ‘the’, respectively; they are much more likely to produce /a/ or /ðe/ (I’ll ignore the dental plosive they’d use for /ð/ so as not to loose focus here), both of which would be intelligible provided they were not stressed. In short, we need to get learners to focus on getting the stress on the right words in an utterance, and also on getting the sounds right in these words (right as per the LFC). At the same time, as teachers we need to learn to be more relaxed about the exact vowel qualities that learners produce in words that would be given a weak form pronunciation by a native speaker in colloquial speech.
Turning to Question 2, I hope it is now clear that in terms of ELF intelligibility, “that” pronounced as /ðat/ because of L1-transfer from Spanish, is just as intelligible in ELF as the RP is /ðæt/. These slight differences in quality are exactly what Jenkins refers to when she says that vowel quality is not important (as long as qualities are consistent), and is not therefore included in the LFC.
(PS. If anybody who reads this can email me and tell me what I have to do to receive warnings about the posting of comments, I’d be really grateful. At my age I’m not a digital native, and I am currently finding that a comment has been posted much later than is desirable, making a response or discussion mostly irrelevant. This is not my desire.)